Overcharging cells at 3.4V per cell

A friend of mine is building his first lifepo4 pack for the house bank on his boat. I've built many so I'm helping out.

He's come across one strange question that I can't answer. This link seems to imply that you can still overcharge a cell even if the voltage per cell is set to just 3.4V: Marine lithium batteries in operation | Nordkyn Design


Now in my 17 years building ebikes and then my electric boat I've never heard of this, and quite frankly can't see anyone being able to harm a cell at 3.4ish volts per cell, but I know there are people on here much smarter than I am, so I'd love to hear your thoughts.

In my opinion (it's just an opinion), the author is substituting hyperbole with fact.

I think he is (correctly) looking at the discharge curve of LiFePO4 and explaining that the rise from 3.4v to 3.65v (the widely-accepted max charge voltage for LiFePO4) is so very steep that it's easy to happen quickly and risk damaging the cell. When you're dealing with 1P, the risk is much less: if I'm charging a single cell by itself and I set the max charge voltage to 3.6, the cell is theoretically unable to achieve a voltage higher than 3.6 (theoretically, because equipment isn't perfect, the real world isn't a lab, ect).

The problem comes when we are dealing with larger LFP packs and cells in series. For example, let's take 16s, a common configuration for most 48v applications. If we want to "play it safe" and charge to only 3.45v per cell, we would want to set a max charge voltage of 55.2v. And if our cells are top-balanced, we would likely see all cells reach 3.45v without much deviation. However, referring once again to the steep discharge curve:


The rise in voltage from 3.45 to 3.65 will be sharp and sudden, even without much capacity added. The problem is that even with well-matched cells, the sharp rise won't happen at the same rate for all 16 cell groups. To put it practically, if you have 16s pack charged to 55.2v, at 3.45v per cell, and you want to charge it higher, perhaps you might want to charge it to 56 volts even. On paper, that would be 3.5v per cell. But because of the sharp curve, it would be entirely possible for cells 3 and 14 to quickly rise to 3.65v, the other 14 cells to only go up to 3.47v, and the max voltage would still only be 55.8v, and your charger would still be trying to add current to the pack! Even if they're well balanced. If you don't have a BMS in place to cutoff charging due to individual-group voltage cutoff, then yeah, you can easily overcharge individual cells in a pack when you try to charge to more than 3.4v per cell.

Hey, ask me how I know!

This is why I say while the author's wording is confusing and potentially misleading, their sentiment is the same as mine, and well-meaning: if you only charge your pack to 3.4-3.45v per cell, the risk of overcharging and damaging your cells significantly decreases. In addition, his sources and others pretty definitively show that since there's barely any capacity to be had over 3.4v anyway, there's really not much reason to charge your cells higher than that. It's a good example of "just because you can, doesn't necessarily mean you should"

Anyway, that's my two cents. Hope that helps.

I have a 220ah 4s lifepo4 pack. I have been charging it to 14.6 volts (3.65 volts) most of the time for the past 5 years, it seems to work for me. Battery is still working great. I never encountered any memory effects. I can monitor each of the cell voltage and also have active balancers to keep the pack balanced.
Without active balancers I wouldnt be able to charge to 14.6 volts, some of the cells always reach 3.65 quicker than the rest activating the BMS. I would probably have to slow charge at lower voltage.
I can see where a pack without active balancers when charged to lower voltage 13.6 volts (3.4 volts per cell) some of the cells might get overcharged if pack went out of balance. I saw that on my pack even at slow charge rates when I tried to charge with a regular charger. Without the active balancers I had to use a balance charger to charge my pack and that took forever on a 220ah battery.
If the battery is in perfect balance, charging at 3.4 volts should never be an issue. The only issue it will take a long time to charge.


Picture of 4s battery. Without the active balancers one of the cells would be closer to 3.6 volts even at the modest 4 amp charge rate.

jonyjoe303 said:

Without active balancers I wouldnt be able to charge to 14.6 volts, some of the cells always reach 3.65 quicker than the rest activating the BMS. I would probably have to slow charge at lower voltage.

Click to expand...

What's different about the cells that reach 3.65 quicker, internal resistance? If so, do they have higher or lower internal resistance?

Zambam said:

What's different about the cells that reach 3.65 quicker, internal resistance? If so, do they have higher or lower internal resistance?

Click to expand...

Could be a lot of things different. Internal resistance, calendar life, cycle life, even variations in temperature, if its a physically large pack. Maybe one side of the pack is facing the sun, you know?

Also, could be nothing different about them, but rather, they're just not perfectly top balanced. Really really close top balance, but not perfect, would still result in some cells starting that steep charge curve before others. I've had LiFePO4 packs that were 0.001v deviation, between 3.2 and 3.45v. Above 3.45v, the deviation increases exponentially.

harrisonpatm said:

Could be a lot of things different. Internal resistance, calendar life, cycle life, even variations in temperature, if its a physically large pack. Maybe one side of the pack is facing the sun, you know?

Also, could be nothing different about them, but rather, they're just not perfectly top balanced. Really really close top balance, but not perfect, would still result in some cells starting that steep charge curve before others. I've had LiFePO4 packs that were 0.001v deviation, between 3.2 and 3.45v. Above 3.45v, the deviation increases exponentially.

Click to expand...

Do you know when a LiFePo4 cell reaches 3.40 V, is it still under constant current phase? or has it switched to constant voltage phase?

It depends on the quality of the cell headway cells don't balance as well as A123 in my opinion. When used in high wattage e-bikes.
Still I like 3.45v ,- 2.9v. of course this is my opinion. But lifepo4 can survive at a higher charge like 3.7v but doesn't like to stay there and we'll settle to 3.5 volt usually. Then under light load it'll settle in the 3.4v range.
How big is this house/boat battery amp hour wise ?

Zambam said:

Do you know when a LiFePo4 cell reaches 3.40 V, is it still under constant current phase? or has it switched to constant voltage phase?

Click to expand...

Batteries aren't what reach constant current phase. Chargers do, or don't. A battery is dumb, it does nothing but respond to the voltage that it is exposed to. Which is why i suggested in a earlier post that you simply charge your pack at 3.4 or 3.45v per cell to avoid overcharging issues.

What's different about the cells that reach 3.65 quicker, internal resistance? If so, do they have higher or lower internal resistance?

Click to expand...

I was using 32650 cells and this was before I had an IR tester. I didnt do a capacity test or IR test since the cells were too large to fit in my opus. The cells were new straight out of the box and decided to use them as they are. I suspect some of the cells might have high IR.
Today I wouldnt use any cells in a pack unless I checked capacity and IR, even if they are brand new.

32650 lifepo4 cell 5500 mah next to the smaller 18650

Clue there are resell wholesalers that have high quality cells for cheap some are older than others but some chemistries last a lot longer than other chemistries and how they're stored in the way they're treated. ???

The author says that 3.4V is enough to obtain near 100% capacity, but that limiting voltage isn't enough to prevent overcharging without also compromising charging. By definition, overcharging is when max capacity has been reached, but the battery is still being charged, so I'm curious how you can overcharge a battery while simultaneously never reaching max capacity. If you never reach max capacity, then the charge current flow must be equal to or less than the internal resistance dissipation rate..?